Holocene-Scale Fire Dynamics of Central European Temperate Spruce
Total Page：16
File Type：pdf, Size：1020Kb
Quaternary Science Reviews 191 (2018) 15e30 Contents lists available at ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev Holocene-scale fire dynamics of central European temperate spruce- beech forests * Vachel A. Carter a, , Alice Moravcova a, Richard C. Chiverrell b, Jennifer L. Clear c, d, Walter Finsinger e, Dagmar Dreslerova f, Karen Halsall b, Petr Kunes a a Department of Botany, Faculty of Science, Charles University, Prague, Czech Republic b Department of Geography and Planning, University of Liverpool, Liverpool, United Kingdom c Department of Geography and Environmental Science, Liverpool Hope University, Liverpool, United Kingdom d Department of Forest Ecology, Faculty of Forestry and Wood Science, Czech University of Life Science, Prague, Czech Republic e Palaeoecology, ISEM (UMR 5554 CNRS/UM/EPHE), University of Montpellier, Place E. Bataillon, Montpellier, France f Institute of Archaeology of the Czech Academy of Sciences, Prague, Czech Republic article info abstract Article history: This study investigated the long-term role and drivers of fire in the central European temperate spruce- Received 7 February 2018 beech forests from Prasilske jezero, Czech Republic. The results illustrate the complex relationship be- Received in revised form tween broad-scale climate, vegetation composition, and local human activities on fire throughout the 27 April 2018 Holocene. Biomass burning was the highest (average 3 fires/1000 years) and most severe during the early Accepted 1 May 2018 Holocene when fire resistant taxa (Pinus, Corylus and Betula) dominated. Using a Generalized Additive Model to assess the response of dominant canopy taxa to changes in biomass burning and fire severity, response curves demonstrate a positive relationship (p < 0.01) between fire resistant taxa and increases Keywords: Climate in biomass burning. Norway spruce (Picea abies) established ~10,000 cal yr BP and expanded during peak Fire biomass burning. Response curves show a slight negative relationship with Picea and increasing biomass Holocene burning, and a positive relationship with increasing fire severity. This suggests that central European Macrofossils spruce forests may not be significantly impacted by fire. Regional biomass burning dramatically Paleoecology decreased with the expansion of fire sensitive taxa (e.g. Fagus sylvatica) ~6500 cal yr BP, yet no dramatic Pollen reduction in local fire frequency occurred. This suggests either human activities or rare fire-promoting Sedimentary charcoal climatic events were important in shaping local fire regimes. Fire activity peaked (6 fires/1000 years) ~2500 cal yr BP and paralleled increases in anthropogenic pollen indicators. Fagus response curves il- lustrates a negative (p < 0.01) relationship with increasing biomass burning and fire severity suggesting that natural Fagus forests may be increasingly vulnerable to projected increases in wildfire occurrence. © 2018 Elsevier Ltd. All rights reserved. 1. Introduction still exist (Feurdean et al., 2012; Feurdean and Vanniere, 2017). More important is the lack in understanding how forest canopy Fire is an important disturbance agent driving changes in taxa will respond to changing fire regimes as a result of vegetation composition, ecosystem structure and function, and anthropogenically-induced climate change. Temperate forest fires nutrient cycling (Boerner, 1983; Bowman et al., 2009; Carcaillet in central Europe are often considered as a negligible ecosystem et al., 2002; Whitlock et al., 2003). Paleoecological studies have disturbance (Adamek et al., 2015) because of the assumed low documented the vital role of climate, vegetation, and human ac- flammability of deciduous forests (Ellenberg, 1982). Yet, over the tivities on global fire activity over millennia (Marlon et al., 2013; past decade emerging literature has demonstrated that fires have Power et al., 2008). Yet, knowledge gaps pertaining to the long- occurred for millennia in central European forested ecosystems term role of fire in particular regions, specifically central Europe, (e.g. Clark et al., 1989; Bobek et al., 2018; Niklasson et al., 2010; Novak et al., 2012). However, crucial parameters of fire regimes i.e. fire frequency and drivers of biomass burning, are neither fully understood nor discussed (Niklasson et al., 2010), specifically from * Corresponding author. central European Norway spruce (Picea abies) and European beech E-mail address: [email protected] (V.A. Carter). https://doi.org/10.1016/j.quascirev.2018.05.001 0277-3791/© 2018 Elsevier Ltd. All rights reserved. 16 V.A. Carter et al. / Quaternary Science Reviews 191 (2018) 15e30 (Fagus sylvatica) forests. granite (Mentlík et al., 2010; Pelc and Sebesta, 1994). Climate models predict more frequent and extreme climatic The surrounding vegetation is dominated by Norway spruce events such as heat-waves and droughts throughout Europe with minor components of European beech, rowan (Sorbus aucu- (Rajczak et al., 2013; Seneviratne et al., 2012), which can elevate fire paria L.), silver birch (Betula pendula Roth), sycamore maple (Acer risk and impact on vegetation (Camia et al., 2017; Lavalle et al., pseudoplatanus L.), and silver fir(Abies alba Mill.). The understory is 2009; Linder et al., 2014). For instance, as a result of increasing dominated by common mountainous spruce forest vegetation such temperature and drought, the susceptibility of mountain plant as grasses (Avenella flexuosa L., Calamagrostis villosa J.F. Gmel., communities to mortality has increased (Allen et al., 2010). Because Luzula sylvatica Huds.), herbs (Prenanthes purpurea L., Senecio ova- extreme climatic events also contribute to an increase in fire risk, tus Willd., Soldanella montana Willd., Trientalis europaea L.), and composition and structure of particular plant communities may small shrubs (Rubus idaeus L., Vaccinium myrtillus L., V. vitis-idaea adapt to include more thermophilic species (Gottfried et al., 2012), L.). Sedges and shade-tolerant species (Carex canescens L., thus favoring more fire-prone ecosystems. As future projections C. echinata Murray, Juncus effusus L., J. filiformis L., Oxalis acetosella suggest an increase in fire risk in central European ecosystems by L.) occur in shady and mesic areas along the inflowing stream, the end of the 21st century (Lung et al., 2013), it is important to Jezerní potok. Ferns (Athyrium distentifolium Tausch ex Opiz, understand long-term fire dynamics in order to determine how Blechnum spicant L., Dryopteris dilatata Hoffm.) and ground pine increasing wildfire activity may impact future temperate forests. (Lycopodium annotinum L.) are also common in the understory Fire activity is determined by both top-down (e.g. climate) and vegetation. bottom-up (e.g. vegetation and human activities) drivers which Based on >50-year long meteorological data (Czech Hydrome- combine to create different fire regimes across varying spatial and teorological Institute) from the closest weather station, Churanov temporal scales. Climate variability is considered to be the domi- (Fig. 1), the modern Sumava region is characterized as a semi- nant top-down driver of fire through its influence on broad-scale humid continental climate with wet and cold winters, and wet energy budgets and variations in moisture and temperature, and mild summers. Interpolated mean annual temperature and À which typically results in regional to continental-scale synchroni- precipitation average 4.5 C and 941 mm year 1. zation of fire activity (Falk et al., 2011). Local factors such as topography (i.e. slope and aspect) and fuel type (i.e. vegetation 3. Methods composition) typically create different mosaics of fire severities (Falk et al., 2011). However, human land use also significantly in- 3.1. Core retrieval, sediment limnology and radiocarbon dating fluences local fire regimes (Gill and Taylor, 2009), yet estimating the extent and magnitude of human-caused fires and land-use activ- In August 2015 a 2.18 m sediment profile comprising of one ities on natural fire regimes has proven to be difficult (see Kaplan gravity core (PRA15-2GC) and two parallel and overlapping long- et al., 2016). Because top-down and bottom-up factors vary cores (PRA 15e2e1 and PRA 15e2e2) were collected from the spatially and temporally (Courtney Mustaphi and Pisaric, 2013; deepest (14.8 m) part of Prasilske jezero from a floating platform. Gavin et al., 2003; Gedalof, 2011), long-term fire histories are The sediment-water interface was collected using a gravity corer necessary in order to investigate the relationship between drivers (Boyle, 1995), and the longer profiles with a Russian corer of fire and their influence on local fire regimes (Whitlock et al., (1.5 Â 0.075 m). Sediments were taken back to the University of 2010). Liverpool where they were subsampled at high-resolution As the role of fire in central European temperate forests is often (contiguous 0.5 cm intervals). Samples were then shipped to deemed unimportant, there is a clear lack of understanding in: i) Charles University where the paleoecological analyses were the relative importance and role of fire in these forests, and ii) the conducted. relationship between the dominant forest canopy taxa and Age-depth relationships were established using ten 14C and a changing fire regimes (i.e. changes in biomass burning and fire 210Pb series (Appleby, 1978)(Table SI1). Age-depth relationships severities). To fill these gaps, we present a ~11,500-year high- were modeled in a Bayesian framework using ‘BACON’